Questions about PID compact temp and cyclic interupt

[Patrix]

Hi everyone,

I have a conventional rubberpress with a upper and under heating plate.
I'm using a 1510SP-1 from siemens

Now i have some issues regarding the PID and the cyclic interupt OB

1. I have two PID compact temp controllers for heating the plates, can they
both be in one cyclic interupt OB or in two separated cyclic interupt OB's.

2. How to determine the cycle time for a cyclic interupt
Current cycle time of CPU:
Shortest = 0.973 ms
Current = 1.3 ms
Longest = 15.8 ms
Cyclic interupt set to 5000 us (5 ms). Is this correct or schould it be faster?

3. After tuning the PID everything works like a peach for a few hours, but suddently the PID controler starts heating and the temperature rises way beyond the given Setpoint.

Errorbits #0000_0800 (Sampling time error: PID_Temp is not called within the sampling time of the cyclic interrupt OB). What does this mean?

The Input warning High wont come TRUE. High limit 250 degrees, input 300 degrees. Is this because of the given error?


Anyone here that could answer these questions?

Thanks in advantage

 

[dcp325]

Hi,

1. Put both in separate cyclic interrupts and use the phase offset of the interrupt to prevent them interrupting at the same time

2. This would depend on the reaction time of your process. For example I've done a IR-heating project with 4 PIDs and 4 interrupts. OB30: cyclic interrupt of 300ms and phase offset of 0ms, OB31: cyclic interrupt of 300ms and phase offset of 75 ms. This resulted in an interrupt every 75ms (different PIDs every time).

3. This error is probably because your PID takes longer than 5ms to process

 

[Patrix]

thank you for the explanation

 

[Peter Nachtwey]

5ms is much faster than necessary for most temperature control applications. A lot depends on the time constant of the system. If the time constant is 1 second a scan time of 100 ms should be fine. Most temperature systems have time constants in the minute range but a rubber press could be faster.

 

[Patrix]

Thanks for the explanation Peter.

 

[A_G]

3. After tuning the PID everything works like a peach for a few hours, but suddently the PID controler starts heating and the temperature rises way beyond the given Setpoint.

Errorbits #0000_0800 (Sampling time error: PID_Temp is not called within the sampling time of the cyclic interrupt OB).

I'm having this exact same problem. Wondering how the OP managed to fix it. I'm fairly new to Siemens. Any help is appreciated! Thanks in advance.

 

[Ken Moore]

I think the PID loop was running faster than the I/O updates. What is the time of your PID, and of your I/O?

 

[A_G]

I think the PID loop was running faster than the I/O updates. What is the time of your PID, and of your I/O?

Thanks for your reply. The cyclic interrupt time is 100000us (1 sec). What do you mean by the time of the IO?

 

[Ali_sh]

Hi everyone,

I have a conventional rubberpress with a upper and under heating plate.
I'm using a 1510SP-1 from siemens

Now i have some issues regarding the PID and the cyclic interupt OB

1. I have two PID compact temp controllers for heating the plates, can they
both be in one cyclic interupt OB or in two separated cyclic interupt OB's.

2. How to determine the cycle time for a cyclic interupt
Current cycle time of CPU:
Shortest = 0.973 ms
Current = 1.3 ms
Longest = 15.8 ms
Cyclic interupt set to 5000 us (5 ms). Is this correct or schould it be faster?

3. After tuning the PID everything works like a peach for a few hours, but suddently the PID controler starts heating and the temperature rises way beyond the given Setpoint.

Errorbits #0000_0800 (Sampling time error: PID_Temp is not called within the sampling time of the cyclic interrupt OB). What does this mean?

The Input warning High wont come TRUE. High limit 250 degrees, input 300 degrees. Is this because of the given error?


Anyone here that could answer these questions?

Thanks in advantage

hi
I have the same problem, please help me